Installation script for on-premise, single server with custom directory

Hopsworks
#1

I am evaluating Hopsworks and would like to install it on-premise on a Ubuntu VM.
Also following

wget https://raw.githubusercontent.com/logicalclocks/karamel-chef/2.4/hopsworks-installer.sh
chmod +x hopsworks-installer.sh
./hopsworks-installer.sh

I got the output

You appear to have following setup on this host:
* available memory: 62
* available disk space (on '/' root partition): 6
* available disk space (under '/mnt' partition): 
* available CPUs: 16
* available GPUS: 0
* file system: ext4
* your ip is: xxx.xxx.xx.xxx
* installation user: bioinformatic
* linux distro: Ubuntu
* version: 20
* cluster defn branch: https://raw.githubusercontent.com/logicalclocks/karamel-chef/2.4
* hopsworks-chef branch: logicalclocks/hopsworks-chef/2.4

WARNING: We recommend at least 60GB of disk space on the root partition. Minimum is 50GB of available disk.
You have 6 space on '/', and no space on '/mnt'.

It appears to me that the installer will install on the default directory /, I have 140GB in a different mnt /data

/dev/sdb1                          147G   61M  140G   1% /data

Is there any simple and quick way to tell the installer to use the custom directory /data instead of the default location / ?

I am a data scientist and very new to the hopsworks, I really appreciate if you can give me a step by step instruction to change the default install location to /data. Thanks so much for your help in advance.

#2

Hi @Yingding

The default installation directory of Hopsworks is under/srv, specifically /srv/hops and /srv/hopsworks-data.

The easiest way to install it on the larger volume would be to either mount the volume under /srv or create a symlink of /srv to /data

#3

sudo ln -s /data /srv/hops
then
./hopsworks-installer.sh

#4

Thank you so much for your quick response @Jim_Dowling @Theo